Background: This study was carried out to examine the association of sub-optimal dialysis initiation of peritoneal dialysis (PD) with all the possible outcomes on PD using survival analysis in the presence of competing risks.
Methods: This was a retrospective cohort study based on the data of the French Language Peritoneal Dialysis Registry. We analysed 8527 incident patients starting PD between January 2002 and December 2010. The end of the observation period was 01 June 2011. Times from the start of PD to death, transplantation, transfer to haemodialysis (HD) and first peritonitis episode were calculated. The sub-optimal dialysis initiation was defined by a period of <30 days on HD before PD initiation.
Results: Among 8527 patients, there were 568 patients who started PD after <30 days on HD. There were 6562 events: 3078 deaths, 2136 transfers to HD, 1348 renal transplantations. When using a Fine and Gray model, sub-optimal dialysis start, early peritonitis and transplant failure were associated with a higher sub-distribution relative hazard of technique failure. There was no association between the sub-optimal dialysis start and the sub-distribution hazard of death or transplantation. In the multivariate analysis using a Fine and Gray regression model, the sub-optimal dialysis start was not associated with a higher sub distribution relative hazard of peritonitis.
Conclusions: Sub-optimal dialysis initiation is neither associated with a higher risk of death nor with a lower risk of renal transplantation. Sub-optimal PD patients had a higher risk of transfer to HD.
Keywords: peritoneal dialysis; unplanned dialysis.